claude

apple-dev

Complete iOS/macOS/iPadOS development workflow plugin for Claude Code — design parity, visual audit (per-View screenshots, light/dark, macOS + iOS), runtime feature verification (real e2e, not mock), and window-shot driving

1 on GitHubby Norvynv1.25.0updated 5 days agodevelopment
/

Plugin apple-dev from marketplace indie-toolkit in n0rvyn/indie-toolkit. Commands read from the repo's manifest — 4 days ago.

This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.

What it installs

Skills (12)

  • apple-swift-context

    Internal context loader for Apple development rules — build cycle, concurrency, UI patterns, plan-execution principles — read section-targeted from references/apple-swift-rules.md. Not user-invocable; loaded when another skill calls it explicitly (dev-workflow:fix-bug does, for Swift/iOS bugs). NOTE

  • asc-listing

    Everything on the App Store Connect backend: what to put in each box, and what the backend actually holds right now. Use when the user says 'ASC listing', 'asc 上架材料', 'ASC 填写', 'app store listing', 'privacy labels', or asks to read the live state — 'ASC 现在填的是什么', '关键词字段实际是什么', '我改的 ASC 字段存进去了吗', '提交

  • asc-submit-preview

    Use before submitting to App Store, or when the user says 'asc submit preview', 'asc pre-submit check', '上架前自检', 'will this pass review'. Checks app code against Apple's App Review Guidelines to catch common rejection reasons. Not for ASC 后台材料 (privacy labels / screenshots / store description) — use

  • aso-research

    Use when optimizing an App Store listing for search — the user says 'ASO', 'ASO 优化', '关键词优化', 'app store 关键词', '搜不到我的 app', 'why doesn't my app rank', '商店排名', 'keyword research', or asks whether a trademark can go in the app name/subtitle/keywords. Pulls REAL data from Apple's own endpoints (live me

  • characterization-test

    Swift/iOS/macOS only. Use before refactoring legacy Swift code, or when the user says 'characterization test', 'lock behavior', '特征测试', 'write characterization tests', 'snapshot behavior'. Generates Swift Testing (`@Test` / `#expect`) cases that capture current behavior as a safety net before refact

  • design-parity-build

    Use when auditing a Claude Design (or similar handoff) against an iOS codebase, or the user says 'design parity', 'design 1:1', 'verify Claude Design', 'Claude Design audit', 'audit design implementation', '设计落地审计', 'Claude Design 对齐', '1:1 还原 Claude Design', or '/design-parity-build'. Writes audit

  • generate-design-system

    Use when the user needs to generate SwiftUI Design System code from token specifications, or says 'generate design system'. Creates Apple HIG-compliant design tokens and component styles.

  • mac-app-shot

    Use when you need to VISUALLY verify a running macOS GUI app from the CLI: capture just ONE app window (not the whole screen) by its window id, optionally drive the app via clicks + keystrokes, and read the PNG with the Read tool to check layout / style / behavior empirically. Covers the edit → rebu

  • project-kickoff

    Manual /project-kickoff only; not model-invocable. Runs the project kickoff flow to clarify requirements, converge scope, and produce execution recommendations. When the product's core value rests on measurement/inference accuracy, an efficacy claim, a state-of-the-art algorithm, or an established s

  • render-preview

    Internal tool-wrapper that renders a SwiftUI #Preview to a downsampled PNG and returns a structured { channel, pngPath, downsampled, error } result. Internal callee for run-phase's visual feedback step (hidden from the / menu); not for ad-hoc user requests. Primary path: Apple Xcode MCP RenderPrevie

  • runtime-feature-verify

    Use to verify a running app's FEATURES actually WORK end-to-end — drive each feature's real action and judge it by a REAL (non-mock) signal (clipboard / on-screen result / state change / persistence-after-reopen / real-LLM output / app log), then produce a feature→judge→evidence→verdict table. This

  • swiftui-visual-audit

    Systematically screenshot EVERY View / Sheet / Modal / sub-view of a SwiftUI app — macOS AND iOS (incl. iPad on the simulator or a real device) — in BOTH light and dark via a throwaway XCUITest harness, then score each render against a general UI rubric (pairs with the refactoring-ui skill) to produ

Subagents (7)

  • app-review-auditor

    Read-only App Store rejection-risk audit of an Apple project's code, in one of two modes. mode=guidelines: App Review Guidelines sections 1-5 checked against code, including the 5.1.x App Privacy evidence table and the privacy-policy.md consistency check. mode=claims: every sentence in the store des

  • apple-reviewer

    Use this agent when the user requests a code review of Swift or Apple platform (iOS/macOS/watchOS/tvOS) code. Only use when the target files are .swift or the project is an Xcode project. Do NOT use for non-Apple codebases (TypeScript, Python, Rust, Go, etc.). Performs read-only review for quality,

  • design-parity-auditor

    Runs the deterministic design detectors, then audits a design handoff (Claude Design or similar, already saved to files by the caller) against an iOS codebase: DS / Tokens parity matrix, page inventory, coverage matrix, page-by-page match report and an 11-field Gap List, following design-parity-temp

  • design-reviewer

    Performs design quality review (visual hierarchy, color, spacing) of SwiftUI files. Fresh context — evaluates visual quality independently of implementation decisions. Examples: <example> Context: User completed a SwiftUI page and wants visual polish feedback. user: "Review the design of my settings

  • feature-reviewer

    Performs product + UX completeness review from user journey perspective. Fresh context — validates feature completeness against user scenarios independently. Examples: <example> Context: User completed implementing a feature with full user journey. user: "Review the expense tracking feature for comp

  • render-auditor

    Scores a batch (~20) of ALREADY-CAPTURED SwiftUI screenshots (PNG) against the refactoring-ui Part B rubric (H/S/T/C/D/F), plus raw localization keys, dead controls and crash logs. Returns a severity-tagged gap list per View + appearance and the shots it could not judge. Dispatched by apple-dev:swif

  • ui-reviewer

    Performs UI + UX compliance review of SwiftUI files. Use when run-phase completes a phase that modified UI files. Fresh context review — no implementation memory. Examples: <example> Context: User finished modifying several SwiftUI views. user: "Run a UI review on the views I just changed" assistant

Hooks (8)

  • SessionStart

    bash ${CLAUDE_PLUGIN_ROOT}/hooks/check-xcode-docs.sh

  • SessionStart

    bash ${CLAUDE_PLUGIN_ROOT}/hooks/cache-booted-sim.sh

  • SessionStart

    bash ${CLAUDE_PLUGIN_ROOT}/hooks/cache-device-udid.sh

  • PreToolUse

    bash ${CLAUDE_PLUGIN_ROOT}/hooks/protect-pbxproj.sh

  • PreToolUse

    bash ${CLAUDE_PLUGIN_ROOT}/hooks/protect-preview.sh

  • PreToolUse

    python3 ${CLAUDE_PLUGIN_ROOT}/hooks/swift-rule-detectors.py

  • PreToolUse

    python3 ${CLAUDE_PLUGIN_ROOT}/hooks/xcodebuild-guard.py

  • PostToolUse

    python3 ${CLAUDE_PLUGIN_ROOT}/hooks/nudge-apple-version.py

View source on GitHub

More from n0rvyn/indie-toolkit

All 6 plugins →

Cross-stack development workflow: brainstorm, planning, phase orchestration, execution, review, debugging, design analysis, knowledge base, commits, and session management

29 skills · 11 agents · hooks

1

mactools

claude

macOS automation toolkit plugin with Notes, Calendar, Mail, Safari, Spotlight, disk reclaim, and more

11 skills · 1 agent

1

Unified plugin lifecycle: create, eval, review, iterate, and package Claude Code plugins and components. Single entry /plugin-master with intent routing.

1 skill · 5 agents

1

Claude Design → Claude Code handoff contract skills (design-spec-contract / flow-navigation-contract / handoff-manifest). Runs INSIDE Claude Design; fetched by repos via raw GitHub URL. NOT a Claude Code marketplace plugin — intentionally absent from marketplace.json.

3 skills

1

WeChat mini program development QA: headless UI verification against running devtools, and pre-submission config/compliance checks

2 skills

1

Similar plugins

superpowers

claudecodex

Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques

15 skills · hooks

obra/superpowers 316,836

context7

claudecodex

Up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your coding agent's context.

1 MCP server

upstash/context7 64,709

Build powerful, stateful CLI interfaces for any GUI application using the cli-anything harness methodology.

5 commands

HKUDS/CLI-Anything 58,665

container

claude

Teaches Claude container's command surface and how it maps to Docker, Lima, Colima, and Podman on macOS

1 skill

apple/container 51,329

Automates browser interactions for web testing, form filling, screenshots, and data extraction

12 skills

vercel-labs/agent-browser 50,373

Use Alibaba Open Code Review from local Codex.

2 skills

alibaba/open-code-review 50,068

Maintain apple-dev?

This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.

claude & codex plugins6 plugins